DATA REDUCTION VERSUS FEATURE SELECTION IN APPLICATION OF DAILY MAXIMUM POWER LOAD FORECASTING

Krzysztof Siwek

ksiwek@iem.pw.edu.pl
Warsaw University of Technology (Poland)

Abstract

Load forecasting task of small energetic region is a difficult problem due to high variability of power consumption. The accurate forecast of the power in the next hours is very important from the economic point of view. The most important problems in prediction are the choice of predictor and selection of features. Two methods of features selection was presented – simple selection using of genetic algorithm and dimensionality reduction methods for creating new features from many available measured data. As a predictor the Support Vector Machine working in regression mode (SVR) was chosen. The load forecasting results with SVR are presented and discussed.


Keywords:

load power forecasting, dimensionality reduction, genetic algorithm, support vector machine

Ashlock D.: Evolutionary Computation for Modeling and Optimization. Berlin, Germany: Springer-Verlag, 2006.
  Google Scholar

Fodor I.: A Survey of Dimension Reduction Techniques. Raport techniczny, 2002.
DOI: https://doi.org/10.2172/15002155   Google Scholar

Gill P., Murray W., Wright M.: Practical optimization. Academic Press, London 1981.
  Google Scholar

Goldberg D.: Genetic Algorithms in Search, Optimization, and Machine Learning. Reading, MA: Addison-Wesley, 1989.
  Google Scholar

Jackson J.E.: User guide to principal components. Wiley, NY, 1991.
DOI: https://doi.org/10.1002/0471725331   Google Scholar

Osowski S., Siwek K., Świderski B., Mycka Ł.: Prediction of power consumption for small power region using indexing approach and neural network. Lecture Notes on Computer Science, LNCS-6352, 2010, str. 54-59.
DOI: https://doi.org/10.1007/978-3-642-15819-3_8   Google Scholar

Osowski S., Siwek K.: Regularization of neural networks for load forecasting in power system. IEE Proc. GTD, 149, 2002, 340-345.
DOI: https://doi.org/10.1049/ip-gtd:20020194   Google Scholar

Sammon J.W.: A nonlinear mapping for data structure analysis. IEEE Transactions on Computers, No. 18, 1969, str. 401–409.
DOI: https://doi.org/10.1109/T-C.1969.222678   Google Scholar

Siwek K., Osowski S., Świderski B.: Trend elimination of time series of 24-hour load demand in the power system and its application in power forecasting. Przegląd Elektrotechniczny, vol. 87, No 3, 2011, str. 249-253.
  Google Scholar

Schölkopf B., Smola A.: Learning with kernels. MIT Press, Cambridge MA, 2002.
  Google Scholar

Vapnik V.: Statistical learning theory. Wiley, NY, 1998.
  Google Scholar

Van der Maaten L., Hinton G.: Visualizing Data using t-SNE, Journal of Machine Learning Research, Vol. 9, 2008, str. 2579-2605.
  Google Scholar

Van der Maaten L., Postma, E.: Dimensionality reduction: a comparative review. 2009, int. report TiCC TR 2009-005.
  Google Scholar


Published
2013-05-16

Cited by

Siwek, K. . (2013). DATA REDUCTION VERSUS FEATURE SELECTION IN APPLICATION OF DAILY MAXIMUM POWER LOAD FORECASTING. Informatyka, Automatyka, Pomiary W Gospodarce I Ochronie Środowiska, 3(2), 9–12. https://doi.org/10.35784/iapgos.1445

Authors

Krzysztof Siwek 
ksiwek@iem.pw.edu.pl
Warsaw University of Technology Poland

Statistics

Abstract views: 183
PDF downloads: 161